[BioC] Defining Weights in marrayNorm.

michael watson (IAH-C) michael.watson at bbsrc.ac.uk
Wed Aug 13 13:22:52 MEST 2003

Hi Guys

Almost finished with this thread, thanks alot for your help :-)

>From the documentation on maNormMain and maNormLoess, I should be able to use something like this to get my weights considered:

 	mydata.norm <- maNormMain(mydata, f.loc = list(maNormLoess(x="maA",y="maM",z=NULL, w=weights)))

where weights is my vector of weights.  However, if I try that, I get:

	Error in model.frame(formula, rownames, variables, varnames, extras, extranames,  : 
        invalid variable type

The documentation states that w should be:

	w: An optional numeric vector of weights.

I have weights as a vector, 19200 long, all ones and zeros.  My marrayRaw objects have 19200 spots.  So I do wonder what could be going wrong, I am sure it is me doing something wrong rather than a bug, however the documentation doesn't really give many clues away :-(

Finally, if I try (without the weights):

mydata.norm <- maNormMain(mydata, f.loc = list(maNormLoess(x="maA",y="maM",z=NULL)))

I get no errors, so it is the weights that are causing the problems!

Sorry to go on about this, I know I should use limma but I really would like to get this working!


More information about the Bioconductor mailing list